Schema reporting tools - XQuery use case?

[Hans-Juergen Rennau]

Hello Lisa,

the article was very interesting to me, thank you! Whereas hitherto I only thought of "schema reports", your work shows me that there is a parallel approach one might call "schema querying". (In fact, I borrowed this term from the namespace URI you used in your article.)  

Trying to define the terms: schema reporting transforms schema documents into new documents conveying part of the schema information (e.g. listing valid data paths); the result is a new resource. Schema querying, on the other hand, is a functionality that provides answers for certain kinds of questions referring to the schema. Schema querying tends to provide dynamic information, referring to instance data, whereas schema reporting usually delivers static information. The border is not sharp: the smaller the report's scope, the closer it resembles an answer to a query, and the use of config parameters may link the report to instance data (e.g. by filtering data). Certainly your function 'load-schemas-by-URI' can be regarded as schema reporting (if you add the possibility to collect the retrieved schemas into a single document).

Now I think that the "important use case" I spoke of comprises both, schema reporting and schema querying. And one thing for sure, schema reports can be a convenient basis for implementing schema querying.

With kind regards,
Hans-Juergen




----- Ursprüngliche Mail ----
Von: Xia Li <http://x-query.com/mailman/listinfo/talk>
An: Hans-Juergen Rennau <http://x-query.com/mailman/listinfo/talk>; http://x-query.com/mailman/listinfo/talk
CC: http://x-query.com/mailman/listinfo/talk
Gesendet: Mittwoch, den 9. September 2009, 10:54:05 Uhr
Betreff: RE:  Schema reporting tools -  XQuery use case?


As a typical use case, schema information is frequently used to process
GML(Geography Markup Language) instance data. Mostly, the schema
querying part is done with the help of Xerces-J XML schema processing
library in Java. But I do think it is quite useful to have the same
thing done in XML processing language directly instead of having to call
external functions. Ideally, it is best done in XQuery processors and
XPath 2 spec does provide some functions to retrieve schema information.
However, the problem is it is not widely supported feature, and there
barely open source XPath processors that support such functionalities. I
developed an XSLT function library to query schema information. It
essentially contains functions to retrieve a type definition for a given
element. 
    
http://www.ibm.com/developerworks/library/x-schemanode/

It is a much more intriguing programming experience to do this in XSLT
than in Java and I think it can be equally done in XQuery as well.

Lisa